Belly Number Research

[JerrryR]

Another one for the collection, an XLH.

67XLH-721X
Belly# 667-2494

Happy Easter to all that celebrate it.
JerrryR

[hennesse]

BUT, if one delete all the info that skews the percieved/hoped-for end result to make the research look better, one opens himself to criticism and rightly so...
I'm just saying the diff between belly serial and engine serial from different year idents should be included in the above data.
One can't pick and choose within these data samples the ones that suits one's train of thought.

Patrick - calm down! I just reported the RESULTS SO FAR. We need a lot more data - there were 7800 XLs produced in 1969 - we have only 6 observations. Not a very good "statistical sampling".

I created an Excel spreadsheet that is OPTIMIZED FOR DATA ENTRY - I simply enter the serial and belly numbers, and it does all the calculations, except for the prior year bellies.

The prior year bellies are a very complex relationship. Consider the following hypothetical situations:

1952 - they produced 5000 bellies, but only 4000 bikes. So we have 1000 extra bellies to be used on next year's bikes.
1953 - they produced 5000 bellies, but only 4000 bikes. So now we have 2000 extra bellies to be used on next year's bikes.

or maybe

1952 - they produced 5000 bellies, but only 4000 bikes. So we have 1000 extra bellies to be used on next year's bikes.
1953 - they produced 3000 bellies, and 4000 bikes. So now we have ZERO extra bellies to be used on next year's bikes.

This is going to take MORE DATA, and a LOT OF THOUGHT, to make some kind of educated guess about the relationship. N'est-ce pas?

All we can do at this point is to make some gross generalizations on the "mainstream bikes" - those with same year serial/bellies.

Here is a scatter chart of the observations, prior year bellies omitted. Each blue dot may represent one, or many, observations

scatter.jpg (52.04 KiB) Viewed 12713 times

What can we make of this? In general, serials and bellies stayed within several hundred of each other. This range increased in 1967-1969, as production soared. There are a few "outliers", but with so few data, we can't be sure they really are outliers.

We'll keep collecting data...
Dave

[dr dick]

Yes, it is possible for an early serial number to have a previous year's belly number, but to me that still show a correlation.

Rightly so.
this has been proven so many times its a non issue.

to this whole idea that the internal parts identification (in this discussion engine cases) somehow relates to vin (registered with the states as a statement of legal origin) does show something----- the efficiency of the supply chain as its related to work in progress.

due to the real supply realities of an oe manufacturer the actual release date of order line item (in our case bikes) can vary drastically from the original projection.
this can happen inside the facrtory or outside.

there were years where not only last yrs bellies were used early on, but instances where the next yrs bellies were used on the previous year.

consider the available yearly production numbers,(the ones so often quoted and accepted as gospel) have been shown time and again to be less than the actual production numbers by direct contact with the factory and the states . read as there is incorrect info is being reported as fact.
if these numbers were fact wouldnt they be solid for all years?
of course they would.
but
yet there are years where no one can site the actual production.
how can that be?

all this is mechanical astrology, nothing is set in stone.
this about internal production deviation.
seriously you wanna track that?
to what end?

hd didnt care about facrory serials vs state serials.
they cared about how many bikes they were able to make and sell.
the correlation of the numbering system be damned if it meant not selling a bike that was on order.

[EKHKHK56]

Dr, some great points. Especially the next year number being used on previous model. Wondered about that. It is interesting if only for novelty putting square pegs in the round holes and it is bringing out some good information on related circumstances. The production sheets are a bit sketchy. I've seen two. Does anyone have a more accurate list or was it lost in the flood? How many 53KKs is one that always comes up. And the elusive but proven 54KHK.

[thefrenchowl]

Hi...

I'm not even convinced the bikes went out the factory in an ordelly fashion, 1001 then 1002 etc till the end of year...

My 54 KHK has one of the last serials for the year, 2540, but is trapdoorless (trap door is supposed to be a mid year change in 54) and has the early moded/reworked/rewelded K trees.

Dave, production was more like 1700 cases and they built 1500 bikes. These extra cases get assembled in the next year, so about January they guess how many more to end the year and start the new model year without itches or go slow in the production area and say "same as last year, lets have 200 extra ones..."

So they don't add up year after year to glut the system.

We also know there wasn't that many D stamped KH cases re-stocked in 57 since pretty fast they had to cast new KH cases, no D and machined partly on KR jigs (race bellies) and partly on Sportster jigs (provision for torrigtons for cams on these with lips in the back, not for bushes)

Patrick

[hennesse]

Another one for the collection, an XLH.
67XLH-721X
Belly# 667-2494

Jerry,

You have either had too much beer - or not enough.

According to Bruce Palmer, How To Restore Your Harley Davidson
the late 1950 - later [ 1969 ] "Crankcase Production Numbers" ["belly numbers "]
the first number followed this code:

1 = E & F series [ 61" and 74" Big Twins ]
2 = S & ST series [ Model 125 and Model 165]
3 = W & G series [ 45 ci ]
4 = not used
5 = K series
6 = B & BT series [ Hummer, Super 10, Pacer, Scat, Ranger, Bobcat ]
7 = XL series [ Sportsters ]

So your bike with belly 667-2494 would have to be something like this:

Ace Elliot's Two Cylinder 1951 Model 125

ace.jpg (87.93 KiB) Viewed 12679 times

I rode Ace's creation back in 2000 at Davenport where I took the photo. First gear was a little cranky. Second smoothed out and pulled hard. Just as I hit third, I had to pull in the clutch and slam on the brakes to avoid mauling pedestrians, vendors, small buildings, etc. It's only 250cc, but I bet it would whip your K-model butt. It's down at Wheels Through Time now, but maybe we can get Dale to let us set up a match race against your 54 KHK...

Also, my computer says that 721X minus anything = NAN ( Not A Number )
Can you provide an actual number?
(It has to be a Real number, not one of those Imaginary Square Root of Negative One thingies)
-----
For those interested in earlier models, Palmer describes the belly numbering scheme used for 1930 to 1947, and the one used for 1948 to mid-1950. And the exceptions.
Don't be a cheapskate - buy his book. You'll never regret it.

[JerrryR]

It appears that for the new model 67XLH motor the factory changed the BN first digit to 6. We have seen other 67Hs that back this up. It also appears that at least by 68 they went back to using the number 7 for the first BN digit. I can't say if this was a running change back in 67 or if it was a clean change with the 68 model year for the H. If anybody has a 67XLH I would appreciate your input on this.

67 XLH 2.jpg (147.52 KiB) Viewed 12677 times

VIN 67 XLH Sized.jpg (155.61 KiB) Viewed 12677 times

67 XLH BN Sized.jpg (152.48 KiB) Viewed 12677 times

[EKHKHK56]

Sounds like Jerry needs to find the correct Hummer cylinders to match! That's a neat bike Dave! I have it in an old Popular Mechanics. Great fit and finish down to the exhaust tips! So 1968...the first one 68XLCH2001 has a 7 for first BN. Than 67 for when built/stamped.

[thefrenchowl]

I rest my case, some factory guys in charge of stamps did not even knew what the belly code meant!!!

Thanks for that gem, Jerry, when I read your 1st post, I thought it was a typo...

I used to have a 67 CH, some 20 years back but never checked the belly... It is somewhere in GB still...

Patrick

[Magneto Sportster]

It appears that for the new model 67XLH motor the factory changed the BN first digit to 6. We have seen other 67Hs that back this up. It also appears that at least by 68 they went back to using the number 7 for the first BN digit. I can't say if this was a running change back in 67 or if it was a clean change with the 68 model year for the H. If anybody has a 67XLH I would appreciate your input on this.

I'm hypothesizing that this was due to the differentiation between XLH and XLCH engine cases that began in 1967. As we all know, prior to 1967, all models used the same cases, but the electric start era changed that in '67. I will check my 1969 XLH when I get home to see if the bellys begin with 6 or 7.